CARY SAS opens its SAS Innovate conference in Florida on Tuesday, and a hot topic of discussion will most likely be the rapid developments in Artificial Intelligence. Reggie Townsend, vice president of the SAS Data Ethics Practice, certainly has a lot to do with AI in the headlines every day, largely due to the rise of ChatGBT and IBM’s warnings that AI will replace nearly 8,000 of the its workers.
In a two-part question-and-answer session, Townsend, who advises the Biden administration on AI and serves on EqualAI’s board to combat bias in AI, talks about the significant issues and possibilities the world needs face.
- Some have called 2023 the year of artificial intelligence, what’s the latest in the field, and why might this designation be accurate?
Artificial intelligence, analytics and machine learning can turn disruptions into opportunities even in the face of geopolitical risks, climate change, supply chain disruptions and economic inflation. We see organizations around the world using analytics and artificial intelligence to make smarter business decisions. In many cases, the success of these efforts is determined by the speed and quality of how their AI and analytics solutions are implemented.
A recent development in AI analysis and deployment is the use of techniques called ModelOps, short for Model Operations. ModelOps takes care of getting analytical models up and running faster so your organization realizes results faster. It is the indispensable technology to quickly and securely implement scalable predictive analytics and AI.
When an organization uses a ModelOps approach, it’s important for them to ask questions about the process that can help them identify bottlenecks, develop a better understanding of where to focus initial attention, and where to make updates.
The ModelOps approach also provides the user with the opportunity at the start of a project to rally the organization to focus on results. ModelOps can improve clarity on the processes involved.
Between new advances like ModelOps, the rapid rise of generative AI like ChatGPT, and AI regulations being developed around the world, the year 2023 looks significant. But, as opposed to just one year, we could see it as the dawn of a new era of AI.
SAS Executive Selected to Join EqualAI Board to Fight Bias in Artificial Intelligence
- Where are the potential points of failure for AI implementations, especially those use cases where AI technology is integrated into supply chain management? What redundancies are built into these systems to avoid failures?
With any model trained on historical data, when something unprecedented occurs, the risk of model failure increases. This is true for supply chains or any system that depends on AI.
But AI can also be the saving grace. For example, prior to the pandemic, Georgia-Pacific, one of the world’s leading pulp and paper producers, implemented a comprehensive data and analytics strategy.
They faced many challenges when it came to speeding up production. Using SAS Viya and our IoT solutions, they reduced the time it takes to build and deploy models by up to 70%.
Then COVID-19 hit and people started hoarding paper products and consumer goods. Georgia-Pacific has seen a 120% increase in demand for toilet paper, tissues and other products. At the same time, there has been a disruption in the global supply chain.
Because Georgia-Pacific already had a robust analytics strategy in place, it was able to scale back its current efforts to overcome this disruption.
They reduced unplanned downtime by 30% and ultimately improved equipment efficiency by 10% to get more products to stores faster. This has resulted in lower maintenance labor costs, less scrap and scrap, and increased production. Additionally, they have 15,000 models running on SAS and are fully prepared for the next outage.
To be better prepared for future disruptions, many companies are harnessing the power of artificial intelligence through simulation and digital twins. These are digital reproductions of real-world systems such as a connected supply chain. These replicas duplicate existing processes with algorithms and IoT connectivity so computers can understand the physical systems involved.
Businesses gain many benefits, including lower costs, by implementing digital twins because they are not limited by the limitations of the physical world. A company can refine product formulation, conduct prototyping and product testing at scale, and optimize resiliency for supply chains faster and less costly than it otherwise could.
The SAS executive will serve as an AI advisor to the Biden administration
- How might the more widespread adoption of AI and AI tools change the U.S. workforce? How might workforce pipelines need to change to prepare future workers for relevant AI and digital skills?
For the widespread adoption of AI to be successful, we need to increase understanding of AI, because there is so much fear and confusion around it. We need to build the fundamental knowledge of AI in the public so that people understand the realistic ways it can help us and the ways it is far less likely to harm us.
We’re already seeing examples of AI, like chatbots, that handle tasks that are easily automated. I think we will see AI become a complementary tool, enabling people to work more effectively, multitask, and focus on work that only humans can do. As impressive as artificial intelligence may be, it still lacks the complex thinking capabilities of humans. And for any workflow that uses AI, humans will need to be in the know to check for bias and fairness, and ensure people aren’t harmed. As you can imagine, these considerations are top priorities for the SAS Data Ethics Practice.
In addition to building fundamental understanding, the lack of AI skills in the workforce inhibits the widespread and effective use of AI. In research published by SAS last fall, 43% of respondents from the US, UK and Ireland indicated that artificial intelligence and machine learning will be top investment priorities over the next one to two years. It was long before proponents of data technology such as data visualization, data analytics and big data. Trouble is, 63% also say the biggest skill shortages are around AI and machine learning.
The survey also indicated that employers are deemphasizing four-year degrees and placing greater value on practical case studies, project work and other relevant training. Industry-recognized certifications, including from technology vendors, were deemed as relevant as college degrees, as were participation in hackathons and data challenges, demonstrating technical, problem-solving and teamwork skills.
To help more people take advantage of the proliferation of AI and close the skills gap, it will take a combination of expanding AI hands-on work in universities, upskilling/retraining people in tech and non-tech roles, empowering employees to do online training or participate in hackathons and grow the data science community. It will also be important to use more modern, open and multilingual tools that will increase data science productivity and enable end users to perform basic analytical tasks, allowing data scientists to focus on core tasks. By democratizing analytics, more people can join the field.
- What are the ethical issues of greatest concern for the development and implementation of artificial intelligence and machine learning tools?
Fundamentally, AI is about automated decision making, so when done with a commitment to fairness, transparency, accountability, and with humans at its core, decision making is beneficial everywhere. Whether it should be automated, and how, is the dilemma. There are obviously areas where AI needs to be carefully scrutinized and carefully regulated. Wherever decisions are made that affect health, well-being, finances and freedoms, we must be careful that artificial intelligence can cause harm on a large scale.
If Amazon recommends a shirt I don’t like, it doesn’t really matter. If someone is denied a home loan because of historically racist data, that’s a serious problem. Or if some populations are underserved by the health care system based on biased data, that is unacceptable. Law enforcement, homeland security, healthcare, and banking are areas where AI risks perpetuating historic injustices, but they also contain opportunities for AI that could actually help people, too.
I just wrote about the balancing act of innovation considering historical data issues in a recent blog, A Call to Action: Empowering Minorities in the AI Revolution.
- What will increased competition mean for AI development in terms of potential risk as private sector and perhaps government competitors scramble to leapfrog each other?
The investments made in artificial intelligence are incredible. It’s an exciting time to be in this business. I am fortunate to be addressing these issues alongside SAS employees, customers and partners, as well as my fellow members of the National AI Advisory Committee and EqualAI’s board of directors. These conversations made clear that we were not going to solve the risk associated with AI with regulation alone. It requires a global approach involving people, processes and technology.
Limits, in the form of regulation, should be placed on the diffusion of AI technology rather than its development. If we limit development overall, there are other countries that will leapfrog and gain potentially insurmountable gains in AI technology.
The regulation will provide the framework and barriers to instill the responsible use of AI. But instilling engagement and consistency will require broad cooperation between AI developers and users. There will still be ample room for innovation but, with clearer guidelines, less risk of unintended harm to society and customers, as well as a company’s reputation, brand and bottom line.
Reducing risk is also a matter of educating organizations and people about responsible AI practices. So many negative outcomes simply result from a lack of awareness of the risks involved. If we can increase the overall knowledge of AI, we can see unintended damage decrease dramatically.
COMING SOON IN PART TWO: What’s happening with AI in SAS
#dawn #era #artificial #intelligence #SAS #executive #WRAL #TechWire